MEDIUM 6.3 NVD

CVE-2026-93840

vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids(). At

vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids(). Attackers can supply token IDs above the output vocabulary that pass validation, causing LogitBiasState to corrupt GPU logits state and allow concurrent requests to sample tokens outside their allowlists.

References

Published: 2026-09-18 · Source: NVD · Feed updated: 2026-09-18
This medium severity vulnerability with a CVSS score of 6.3 was published on 2026-09-18 via NVD.
vulnfeed aggregates 14357 vulnerabilities from NVD, CISA KEV, Ubuntu, Debian, Red Hat, Kubernetes, Exploit-DB, OSS-Security, GitHub and OpenStack — updated every 4 hours.